AHP Part 1

[@datascientistph]

how you arrived at the last column and last rows?

[@csjoeker]

Thank you lady. You saved my exam result. Much obliged

[@altuush1101]

lol u need to watch the video again

[@salman_3833]

+Altuush Batsaikhan what do you even mean?

[@inglocines]

Thank you for posting this video. It helped me a lot. It made me understand AHP easily.

[@sarun777]

Thank you for the lesson. Very concise and easy to understand.

[@charlesivanjolyousfernande5765]

it really helped me with my thesis. thanks mam.

[@kzy711]

Very clear and easy to understand. Thanks!

[@masouddelghandi7066]

Hi, How do I derive [C] ? ? ?

[@sowmithreddy4546]

any idea on how to calculate the pairwise matrix of criterias?

[@abdurrauf9434]

Kindly please tell me how to find Ws in this... I didn't get it.. Please mam

[@artisticgirl6797]

is R.I value is fixed and from where we are getting this?

[@elleuchmarwa962]

hi , please how can i do the binary comparison after the W calculation ???

[@Ukazi_Biblical_Womanhood]

thank you for the tutorial, it is very helpful. However, who decides upon what value of importance to given a certain criteria over the other. For example; i am doing a project of selecting a new landfill site using certain criteria; how do i decide whether distance from residential areas is important over infiltration rate and by how much?

[@sebcde]

Phindile Penelope I think the point of the method is to bring mathematical structure to personal subjective preferences. So typically the decision maker will fill in the matrix based on his preferences, or based on rational studies. If you are the analyst, you should ask the DMs to rate their preferences, otherwise, fill them yourself :) I'd be curious to hear your thoughts and solutions on the question if you were able to find any.

[@Ukazi_Biblical_Womanhood]

Hi again, pertaining assigning values of preference, i adapted those used in literature for international studies since there aren't any studies in South Africa. However, i am stuck upon checking for consistency of my weights. I checked out 'Part 2' of your tutorial but i was confused how you calculate {Ws}; please assist.

[@Gaestje]

Helpful video, however I am confused in between videos. No clue on how to get the WS (weighted sum) value; what is C?

[@GERRARD2083]

+Nick Beltman well C is the original non normalized vector and Ws is just the product of [C]*[W] were W is the vector representing the average of the rows of the normallised matrix that was calculated.

[@kadirmokrane9848]

Nice presentation and talk. However i'm wondering to know how to estimate ration consistency with Fuzzy-AHP. An example would be valuble. Tks

[@yasirmohyh]

Hello, who is the professor who present this video?

[@kasklore5063]

Who's the speaker ? Any references for the materials used in the video ?

[@GloriaStarns]

The material for this video is based on the presentation on AHP in the text Machine Design, 5th ed., by George Dieter and Linda Schmidt.

[@MonzieBoo17]

What if B is more important than A?

[@rssharma184]

A very helpful tutorial. Thank you so much for sharing. :)

[@AmateurPeanutButter]

Thanks a lot, this explanation is easy to follow.

[@scottdoremus4735]

Must the pairwise comparison always be symmetrical?

[@GloriaStarns]

Hi Scott, The matrices must be square for the math to work correctly-however that may not be what you are really asking. So your question may be more related to the way in which the comparisons are completed. Do you mean, for example, that if A is paired with B and given a value of 2, would it be inherent in the AHP process that B paired with A would be 1/2? Since the comparison is attempting to give relative value to each of the two items being evaluated, it is important that the rankings be consistent. So if A relative to B is 2, then B relative to A must be 1/2. Sorry if I have not really understood your question. Satay's original paper is widely available and I have a pdf of it if you have trouble finding it.

[@scottdoremus4735]

Hi Gloria, Yes, your interpretation of my question is correct. A to B = 2 and B to A = 1/2. While I have your attention...what is the advantage of AHP over say, SMART? It is much more complicated that for sure but, what is the real benefit? In other words, what does AHP capture in decision making that the Simple Multi-Attribute Rating Technique does not?

[@GloriaStarns]

Scott Doremus Hi Scott, I am far from an expert in Decision Analysis methods but a quick search on SMART (so that I would even know what the acronym stands for!) indicated that the distinction may be mostly related to scaling issues. I read a very brief abstract (Scale sensitivity in the multiplicative AHP and SMART, F.A . Lootsma) that suggested as much. Where you have orders of magnitude of differences, it appears SMART may be the better choice. With that said, my expertise on SMART is entirely exhausted. I appreciate being asked--I learned something new today.

[@raghuramkolathu9021]

Hi- thank you for this knowledge. Would you be comfortable to connect sometime on WebEx ? I am trying this for the first and I need some guidance

[@albfsd]

Hi. How are u? I want to use AHP on my research work but I am little confused about its application. Actually I am working on inspection skill of human labour and factors affecting on it. And I want to use AHP to classify the available manpower into three skill levels (Low, Medium, and High) based on the factors that can affect the skill of quality inspectors. Plz guide me can i use AHP for this purpose? if yes then plz give your valuable suggestions and useful material that can be helpful for me to solve my problem

[@morganmckay93]

Im conducting AHP analysis using 6 criteria to evaluate amongst 5 alternatives. There are 5 decision makers that I have asked to compare each criteria against each other for each of the 5 alternatives. How do I incorporate the comparisons from all 5 decision makers looking at all 5 alternatives into the same AHP analysis? Thank you!!

[@sebcde]

Morgan McKay Hello Morgan, I also asked myself this question when watching the video. My guess is that you could either have each DM go through the AHP process, and simply have them compare their individual results in the end and reach a compromise. Or you could somehow average the weights of all DMs and therefore have an average result. Have you found any answers since you posted the question?

[@morganmckay93]

sebcde Hey, yes I took a geometric mean of the decision makers' pairwise comparisons to synthesise into one model as it is the most commonly used method to do so in AHP. Below are sources I used to justify. Hope this helps Ramanathan, R., & Ganesh, L. S. (1994). Group preference aggregation methods employed in AHP: An evaluation and an intrinsic process for deriving members' weightages. European Journal of Operational Research, 79(2), 249-265. Saaty, T. L. (1989). Group decision making and the AHP. In The Analytic Hierarchy Process (pp. 59-67). Springer Berlin Heidelberg

[@lornatwins9944]

I found this very easy to understand. What is your profession. Why do you know this?

[@mruxp0seid0nx32]

Thank you soo much.. you made my task much more easier ... Thank you once again :)

[@BigBird417]

This is for sample only, how about if there are many samples?

[@GloriaStarns]

There are numerous other videos on KZbin that should help you in finding applications other than the one I discuss in the video.

[@meriemboultif6647]

Thank you....this video helped me a lot

[@engemey1]

thank you very much for this very helpful video... I'm just confused and need to ask you is this what they call it Eigen value and Eigen vector method?? or this is a different way!!

[@engemey1]

pleaseeee answer me?

[@GloriaStarns]

engemey1 The eigenvector gives the relative importance of each item being compared; it is the weights resulting from the pairwise comparison. The eigenvlaue itself is used to compute the consistency of the comparisons. There is an excellent tutorial at the following site: people.revoledu.com/kardi/tutorial/AHP/Priority%20Vector.htm. Eigenvectors and eigenvalues are pervasive in the sciences and engineering; if you are a student of science or engineering, I would urge you to explore eigenvalues and eigenvalues more fully. I hope this helps.

[@engemey1]

Gloria Starns it helps a lot thankss again :)

[@kasklore5063]

+Gloria Starns Is there any reference to the material presented in the video ?

[@itchyyboyy]

What is the value used as [C]?

[@dsgdsfhdsghdhdfh]

the original matrix 6:23

[@Ann_dmnhvd]

thank you. you helped me a lot. easy to understand :)

[@dsgdsfhdsghdhdfh]

thanks a lot for explaining this

[@altuush1101]

its amazing thank you very much!

[@intuitiondigression2174]

Thanks for your video. This is good for mechanical application of AHP: you explain the calculation for each passage. However, you do not explain the meaning of each calculation, so interpretation is not easy. For example, why should one sum the element of each column? What does the resultant number represent and why? Also, why should one normalize? And why does averaging by row should provide the importance degree of each criterion? Why does CI has n-1 at the denominator? And what does lambda mean and what does lambda minus n represent?

[@yanjulio]

Thanks so muuuuuch!

[@abdur-rahmanquadri9904]

Thank you...

[@kousikdas6059]

thank you ma'am

[@zakariamarzouk374]

thank you .!!!

[@yashduggad9966]

thank u .... helpfull

[@mamunarrashid12]

Tnx